horizontal scavenger borehole system for plume containment · horizontal scavenger borehole system...

512 Wolkersdorfer, Ch.; Sartz, L.; Weber, A.; Burgess, J.; Tremblay, G. (Editors)

Horizontal scavenger borehole system for plume containmentMartin Holland1, Kai Witthüser,1,2

1Delta-h Water Systems Modelling (Pty) Ltd, Pretoria, South Africa, [email protected]� liated Professor, Institute for Groundwater Studies, University of Free State. Bloemfontein 9300,

South Africa

AbstractWhile scavenger boreholes have been proposed for groundwater remediation for a number of Mine Residue Deposits throughout Southern Africa, a limited number of case studies are available. Furthermore, the use of horizontal scavenger boreholes for site remediation is common in the northern hemisphere but is an emerging technology in South Africa. � is paper presents results and lessons learnt from implementing the � rst horizontal scavenger borehole system for the hydraulic containment of a seepage plume emanating from an unlined Tailings Storage Facility (TSF).Keywords: ICARD | IMWA | MWD 2018, hydraulic plume containment, horizontal scavenger boreholes.

Introduction � e Horizontal Directional Drilling (HDD) Industry in South Africa is largely focused on the installation of underground utilities mainly within the overburden. As a result, di-rectional drilling companies are not familiar with the drilling and equipping of horizontal boreholes for hydraulic plume containment, just as very few hydrogeologists have experi-ence in the design and installation of horizon-tal boreholes. In this case study, two horizon-tal scavenger boreholes were drilled, screened and commissioned to intercept a seepage plume emanating from a Tailings Storage Facility (TSF) and associated Return Water Dam (RWD). � e study area is located in the eastern limb of the Bushveld Igneous Com-plex, 25 km south west of Steelpoort in the Limpopo Province, South Africa. � e area is underlain by massive to poorly layered pyrox-enites and norites of the Rustenburg Layered Suite, intruded by numerous dykes. In-situ weathered and unconsolidated transported overburden covers most of the valley bottom and lower and mid slopes of the valley sides, forming a typical Basement aquifer system. Tailings deposition (from a Platinum Group Metals Concentrator) onto the TSF started in October 2006 and surface- and groundwater monitoring data showed a deterioration of water qualities downstream of the TSF and RWD. Highly elevated nitrate, ammonia, sul-

phate and sodium concentrations in the seep-age water from the TSF contribute to a saline mine drainage chemistry. � e current plume (represented in � g. 1 by sulphate) emanating from the 60 Ha TSF migrates preferentially in zones of deeper weathering along ‘pre-facili-ty’ drainage courses to the east and south-east towards a large perennial river. � e vertical plume migration is limited by the underlying fresh bedrock with signi� cantly lower per-meability compared to the upper weathered overburden with associated higher � ow rates. Results from a site speci� c numerical ground-water � ow and transport model indicated that the plume could be intercepted before reaching the receptors, in this case regarded as the southern tributary and the perennial river to the east. � e sulphate concentration limit for the river is 70 mg/L based on the in-stream � ow requirement. Scenario modelling concluded that two horizontal wells could in theory be more e� ective than 10 conventional boreholes, reducing the Capex and eventually the long-term Opex cost for the predicted minimum pumping rate of 15 years post-clo-sure (expected for 2019).

Methods Following an initial plume characterisation study, additional monitoring boreholes were drilled during the plume interception study to establish sub-surface conditions along the

8_Migitation BOOK.indb 512 9/3/18 12:20 PM

11th ICARD | IMWA | MWD Conference – “Risk to Opportunity”

513Wolkersdorfer, Ch.; Sartz, L.; Weber, A.; Burgess, J.; Tremblay, G. (Editors)

proposed horizontal borehole path. � e drill-ing was preceded by ground geophysical sur-veys, including earth resistivity imaging, and electromagnetic and magnetic traverses adja-cent and downstream of the TSF. � e main aim of the survey was to:• confi rm dyke positions inferred from re-

gional aeromagnetic data,• investigate geological structures and deep

weathering zones as potential preferential � ow paths,

• optimize the selection of drilling sites for scavenger wells.

Ultimately a total of nine geophysical tra-verses were conducted with a total length of 1 918 m. Based on the resistivity survey re-sults, the resistive basement (bedrock) varies in depth from 4 m to approximately 18m be-low surface. Overburden and soil cover varies from absent (rock outcrops) to 1-2 m thick-ness. � e resistivity results were also used to delineate areas associated with deep seepage, indicating saturated conditions immediately downstream of the TSF and RWD.

Boreholes drilled along the proposed hor-izontal borehole paths were used to con� rm

Figure 1 Simulated sulphate plume extent.




the depth to bedrock, likely lithologies and groundwater qualities to optimise the hori-zontal borehole pro� le. � ese observation boreholes also aid in assessing and monitor-ing the drawdown of the water levels dur-ing testing and commissioning. Additional monitoring boreholes were drilled further downstream of the horizontal boreholes to assess the e� ectiveness of the hydraulic con-tainment system. Based on hydraulic test re-sults, the hydraulic conductivity varies from 0.1 to 0.8 m/d for the area with the highest contaminant concentrations (south of the TSF and RWD).

� e information was used to establish a horizontal pro� le to be followed during the horizontal drilling process. � e target depths were essentially the weathered-bedrock inter-face zones perpendicular to the preferential groundwater � ow path. Directional drilling methods use specialized bits to curve bores in a controlled arc which enables bores to be initiated at a relatively shallow angle from the ground surface and gradually curve to hori-zontal (EPA, 1994). While blind horizontal boreholes terminate in the subsurface, in most cases the borehole is turned back to-wards the surface and returns to the ground to form a continuous well (Figure 2).

A summary of the pre-drilling design cri-teria is provided in Table 1, while the initial design considerations are given below:• Th e original aim was to drill the hori-

zontal boreholes from surface to surface along the pre-planned path so that the well screen portion of the borehole is � at or inclined at a speci� ed grade at the de-sired location.

• Once the drill head (with a diameter of 165 mm) has re-emerged at the surface, it is removed from the drill string and the 110 mm OD sleeves (screened and solids) are attached directly behind the drill head and pulled into the borehole.

• A walk-over system consisting of a radio beacon-receiver measures surface loca-tion up to a depth of 16 to 18 mbgl to en-sure the boreholes follow the correct path.

ResultsDrilling was carried out using an American Augers DD-10 percussion rig with an air per-cussion hammer head. Each drill entry point was excavated to around 1 m below surface. Steering of the drill head remained below 2 % per 6 m rod length, which prevented the borehole being completed back to surface. As a result, the two horizontal boreholes were

Figure 2 Horizontal well used to intercept a plume. � e solid line represents a blind well; the solid line plus the dashed line represents a continuous well. (EPA, 1994).

BoreholeDeepest target depth

(m)Length to target

depth (m)Total length (m)

Sulphate (concentrations)

HWELL-B1 18 70140

450

HWELL-B1 9 50 110 96

Table 1 Design criteria of the horizontal scavenger borehole system




completed as blind wells. � e � nal drilled horizontal borehole pro� les, in relation to the closest geophysical pro� le and vertical moni-toring boreholes, is shown in � g. 3.

A� er a horizontal borehole was drilled it was developed using a smaller reamer forcing air and water into the horizontal borehole, � ushing so� sandy overburden material from the open hole. � e drill rods were moved back and forth along the entire screened sec-tion until the discharged water was relatively free of suspended material. � e HDPE sleeve was pushed into the hole with the screened sections installed along the target area. Solid sleeves were used in the upper overburden. � e hole opening with the protruding starter sleeve � tting over the riser HDPE pipe was sealed with a bentonite-based cement block.

� e preliminary pumping tests of the horizontal boreholes included limited step tests and constant discharge tests followed by a recovery monitoring period, mainly to de-termine the potential yields of the boreholes

and to determine the speci� cations for the submersible pumps for the installation. � e detailed schedule of the constant discharge tests is provided in Table 2.

Based on the results, the horizontal bore-holes were equipped with 3-inch submersible pumps which run through an inverter, al-lowing for variable discharge rates. Based on the tests conducted and the limitation of the 3-inch pump diameter, the following ranges are obtainable form the current installation. • HWELL-B1: 0.1 L/s to 0.9 L/s (at 75 m

horizontal distance)• HWELL-B2: 0.1 L/s to 0.6 L/s (at 60 m

horizontal distance)

During the time of the study, the reticula-tion system to the Pollution Control Dam was yet to be installed. Although no formal monitoring programme was implemented yet, the time was used to test the pumps and reticulation up to a transfer dam. Two piezo-metric data loggers were installed within the

Figure 3 Horizontal borehole pro� les in relation to the closest resistivity results.

BoreholeScreened section

(m)Pump intake (horizontal) distance (m)

Constant discharge rate

(L/s)Time (m)

Max. drawdown (m) at pump

intake

HWELL-B1 102 47 0.84 480 4.23

HWELL-B1 66 45 0.54 960 5.05

Table 2 Pumping test summary




observation boreholes closest to the com-pleted horizontal boreholes. Pumping rates were commissioned on preliminary licensed abstraction � gures (established pre-drilling of the horizontal boreholes), namely 0.54 L/s for HWELL-B1 and 0.23 L/s for HWELL-B2. � e drawdown results for the longer-term testing period are shown in Figure 4. No con-tinuous pumping of the horizontal boreholes was possible due to power surges in the sub-station providing electricity to the pumps, resulting in a recovery of water levels dur-ing down-time. � e system is currently be-ing investigated to ensure an uninterrupted power supply to the pumps. While the results indicate drawdowns within the perimeter of the horizontal borehole, the larger zone of in� uence will only be determined once the site-speci� c monitoring programme becomes operational.

Updated post-closure model predictions indicate that pumping should continue be-yond 15 years post closure, until source de-pletion (i.e. reduction of seepage rate once active deposition onto the TSF cedes and the TSF is rehabilitated) and natural attenuation of the remaining plume a� er pumping ceased prevents unacceptable plume concentrations reporting to the receptor (tributary). Figure 5 depicts the seepage plume with a cut-o� value of 70 mg/L a� er 15 years post closure with the commissioning of the two existing horizontal wells. While it appears that the distribution and pumping rates assigned are su� cient to limit the plume migration, it’s only with on-going monitoring and post-audits that model predictions can be veri� ed.

Figure 4 Drawdown results from longer term tests.




ConclusionsHydraulic containment of seepage plumes emanating from an unlined TSF and RWD in the eastern limb of the Bushveld Igneous Complex was required to prevent impacts on a downstream river and ensure legal compli-ance. Detailed geophysical and intrusive in-vestigations as well as numerical modelling assisted in the delineation of zones of prefer-ential weathering and ultimately groundwa-ter � ow and transport pathways. � ese were intercepted using two horizontal scavenger

Figure 5 Cross section showing simulated plume extent 15 years a� er hydraulic plume containment using horizontal boreholes commenced.

boreholes. Current pumping rates are based on licensed abstraction limits, but will be adjusted once a site-speci� c monitoring pro-gramme is initiated to delineate the zone of in� uence and to determine the e� ectiveness of the scavenger system.

References EPA (1994) A lternative Methods for Fluid Deliv-

ery and Recovery. EPA/625/R-94/003. U.S. En-vironmental Protection Agency (EPA). Ohio, United States of America.


horizontal scavenger borehole system for plume containment · horizontal scavenger borehole system...

Documents